A Not So Bullshit Tax Shelter -- Maybe; Its All in the Eye of the Beholder (10/18/13)

Today's blog addresses yesterday's decision from a very good judge, George A. O'Toole, Jr. of USDC MA (Wikipedia here).  The decision is Santander Holdings USA, Inc. v. United States, 2013 U.S. Dist. LEXIS 149441 (D MA 2013), here.  The tax shelter was the STARS shelter -- for "Structured Trust Advantaged Repackaged Securities".  (Where do they get these names from?).  The Judge holds, in effect, that the shelter was not so bullshit and not bullshit at all.  For holdings in the prior cases that the shelter was quite bullshit, see here.

The Santander arrangement was complex, a common feature of bullshit tax shelters designed to deflect attention and/or comprehension.  Judge O'Toole says:

Up close, however, the transaction was surpassingly complex and unintuitive; the sort of thing that would have emerged if Rube Goldberg had been a tax accountant. The government might be forgiven for suspecting that the designers of anything this complex must be up to no good, but that understandable instinctive reaction is not a substitute for careful analysis, and on careful analysis, the government's position does not hold up.

For more discussion of the transaction, I direct readers to the opinion for Judge O'Toole's summary explanation of the transaction (footnotes omitted):

A very brief overview of the transaction is sufficient for present purposes. Sovereign created a trust to which it contributed $6.7 billion of income generating assets. The trustee of the trust was purposely made a U.K. resident, causing the trust's income to be subject to U.K. income taxation at a rate of 22 percent. The trust income was also subject to U.S. income taxation and was attributed to Sovereign, but  with a credit available for the amount paid in U.K. income taxes under section 901 of the Internal Revenue Code ("the Code"). 26 U.S.C. § 901. Sovereign paid U.K. taxes and then claimed credits for the amounts paid in calculating its U.S. income tax liability for the tax years in question. 

The transaction included a number of contrived structures and steps that, each viewed in isolation, would make little or no sense. For example, Barclays had an ownership interest in the trust and as a result received monthly distributions from the trust, which, under the terms of the transaction, it was required immediately to re-contribute to the trust. Standing in isolation, this circular movement of distributions would make no sense. In the context of the entire transaction, however, it was crucial to Barclays' obtaining favorable tax treatment under U.K. law, which gave it the ability to lower its effective lending rate to a U.S. bank. The result of the STARS transaction for Barclays was a net tax gain, which it was able to use to reduce other U.K. tax liabilities that it owed. 

The loan aspect of the transaction was also highly structured in an idiosyncratic way, although it was consistently treated by Sovereign for accounting and regulatory purposes as a secured loan, acceptably to regulating agencies, including the Securities and Exchange Commission and the Office of Thrift Supervision. One feature of the loan arrangements was what was denominated the "bx payment," or the "Barclays payment." It was calculated as approximately one-half of the amount Sovereign paid in taxes to the U.K. on the income earned by the trust. While in the intricacies of the transaction it was actually a monthly credit to Sovereign figured into its interest costs, the government refers to it as an affirmative payment in support of its "effective rebate" argument, and Sovereign accepts that characterization for purposes of this motion.

The issue Judge O'Toole decided was whether, in the objective analysis of whether a payment from Barclays to Sovereign was effectively a rebate of the U.K. tax.  The Court held that it was not a rebate, either actually or in substance.  The net effect of its holding was that the taxpayer gets to claim the full credit and gets to treat the payment from Barclay's in calculating its objective profit from the transaction.

In its decision, the Court says:

Lacking compelling legal authority, the government proffers the learned opinions of its putative expert witnesses. The problem is that their opinions do not matter, because the necessary question is not a question of fact – What happened? – but rather a question of law – How should what happened be classified for purposes of applying the law? That is why this issue is amenable to resolution on a motion for summary judgment. The facts of the transaction are not in dispute. There is no material factual issue about how the credits and debits worked their labyrinthine way through the Goldbergian apparatus. The question is, Should the Barclays payment be treated, as a matter of law, as if it were a rebate from the U.K. to Sovereign? That is a legal question, to be answered by judges, not economists. See IES, 253 F.3d at 351 ("The material facts are undisputed; the question of law before us is the general characterization of a transaction for tax purposes."). 

 * * * * 

Slight as this authority may be, it is enough to outweigh the government's authority for its proposition that a private payment may be recharacterized into a tax item, which is nil. The recent decisions in similar STARS cases do not discuss the issue. See Salem Fin., Inc. v. United States,     Fed. Cl.    , 2013 WL 5298078, at *39-40 (Sept. 20, 2013); Bank of N.Y. Mellon Corp. v. C.I.R., 140 T.C. 15, 40-43 (Feb. 11, 2013). Those cases appear to deal with the question whether the Barclays payment was "in substance" a "tax effect" as a matter of fact, rather than as a matter of law, as I conclude is proper. In other words, they accept the testimony of the government's experts and make a factual finding that the Barclays payment was an effective U.K. tax rebate and consequently a U.S. tax effect. Salem Fin., 2013 WL 5298078, at *40; Bank of N.Y., 140 T.C. at 43. Notably, they do not address the legal question whether a private party payment between Barclays and the relevant bank can properly be classified as a tax effect because it is so much like one in substance, a question that Doyon and the private letter rulings answer in the negative.

Essentially, as I understand it, Judge O'Toole's disagreement with the prior decisions for the Government on the STARS transaction is over whether the substance is consistent with the form.  Judge O'Toole views the substance inquiry as a legal one where he determined that the law treated the substance in this case as consistent with the form.  The other courts, taking a different view, did not treat the law as commanding that the substance was the same as the form.  Rather they looked to the type of expert analysis that Judge O'Toole rejected to say that the substance was not consistent with the form.  Judge O'Toole found support for his analysis in IES Indus., Inc. v. United States, 253 F.3d 350, 353 (8th Cir. 2001) and Compaq Computer Corp. & Subsidiaries v. C.I.R., 277 F.3d 778, 784-86 (5th Cir. 2001), both of which I think are, like Judge O'Toole's holdings, not mainstream holdings in retrospect.  Note that the Tax Court rejected Compaq in the Bank of New York Mellon Corp. STARS transaction.  See 140 T.C. No. 2, fn. 9 (2013) which says:

   n9 We have previously held that foreign taxes are economic costs for purposes of the economic substance doctrine. See Compaq Computer Corp. v. Commissioner, 113 T.C. 214 (1999), rev'd, 277 F.3d 778, 785 (5th Cir. 2001). We are mindful that the Courts of Appeals for the Fifth and Eighth Circuits have subsequently held that foreign taxes should not be taken into account in evaluating pre-tax effects for purposes of the economic substance analysis. See IES Indus., Inc. v. United States, 253 F.3d 350 (8th Cir. 2001); Compaq Computer Corp. v. Commissioner, 277 F.3d 778, 785 (5th Cir. 2001), rev'g 113 T.C. 214 (1999). Nevertheless, the Supreme Court and the Court of Appeals for the Second Circuit have yet to consider the issue, and we are not bound by Fifth and Eighth Circuit precedent here.
   We maintain the position we took in Compaq Computer with respect to foreign taxes in the economic substance context. Economically, foreign taxes are the same as any other transaction cost. And we cannot find any conclusive reason for treating them differently here, especially because substantially all of the foreign taxes giving rise to the foreign tax credits stemmed from economically meaningless activity, i.e., the pre-arranged circular cashflows engaged in by the trust.
   Additionally, excluding the economic effect of foreign taxes from the pre-tax analysis would fundamentally undermine the point of the economic substance inquiry. That point is to remove the challenged tax benefit and evaluate whether the relevant transaction makes economic sense. See In re CM Holdings, Inc., 301 F.3d 96, 105 (3d Cir. 2002).

Although Judge O'Toole's decision grants only partial summary judgement, it looks like the final decision will go for the taxpayer.  The Government is likely to appeal, so that the First Circuit will have the final say.  The validity of these shelters are in the eyes of the beholder, I suppose.  The First Circuit will have differents eyes to behold.  No predictions as to ultimate outcome.  But should a split develop, the worst of all things could happen.  The Supreme Court, which screwed up Frank Lyon, could wade into the mess again, although I suspect that, given that debacle, the Supreme Court may not want to weigh in on that issue again.